Re: lynx-dev New <BR> collapsing patch

[Benjamin C. W. Sittler]

On Mon, 17 Aug 1998, Philip Webb wrote:

> 980816 Dave Eaton wrote: 
> > On Sun, 16 Aug 1998, Philip Webb wrote:
> >> (2) there is no `valid, correct' behaviour, since the specs are broken;
> > I think you can see a number of us dissagree with you
> > that the spec is "broken" and believe current Lynx behavior is "correct",

> it's not clear that anyone other than you disagrees
> the specs are inconsistent & unclear;
> if they do, they should offer some analysis of the specs themselves.

The specification seemed quite unambiguous on this point. Even if most
authors don't understand this, HTML is not a programming
language for controlling visual presentation of documents, it is 
a markup language for hinting at a document's logical structure.

> > though we have no objection to alternate renderings in the code,
> > so long as they are optional, configurable, and not the default.

> i've reviewed the thread(s) & don't see a majority:

>   for keeping current behaviour as default: yourself, LV, MW, NHE (4);
>   for not collapsing <BR><BR> by default: AG, DH, JM, DW, me (5).

Though I'd hardly qualify as a developer, I must agree with the latter
behavior. 

> i'm NOT claiming a majority of lynx-devers NOR suggesting we hold votes,
> but there is a real range of opinion & so far no consensus.

> there's also the little matter of rational argument.
> i presented  4  careful analyses of the HTML 4.0 specs
> & the real-life situation re <BR><BR> & document authors
> (980812 1716 (to which you replied), 980812 2026 (reply to you),
>  980815 1837, 980816 1902):
> you are the only contributor to have replied to my criticism of the specs,
> to which i responded without further comment from you;
> no-one has addressed the real-life state of affairs
> in which the typical document author has very little knowledge of HTML
> & very little time or inclination to learn about its finer points.

> you made a distinction `document structure vs page layout'
> & JM distinguished `procedural/structural mark-up systems',
> which i may understand (are they the same distinction BTW?),
> but if you believe more than a very tiny minority of document authors
> have the slightest knowledge of or interest in such things,
> you simply don't live in the real world:
> even highly professional newspapers haven't time for it,
> eg Washington Post (uses ../ ) & Financial Times (London) ( <BR><BR> ).

> > Now, would it be OK to move on to other topics?

This is my last message on this topic.

> if you've run out of arguments, by all means move on to something else.

> the solution is simple: a run-time configurable choice (needs programming)
> or at least changing the  lynx.cfg  default to match real-life out there.

Run-time configurability is nice, but doing the right thing by default and
allowing this to be changed in lynx.cfg is good enough for me. So far,
I've seen very few documents which rely on multiple <BR> tags for
readability.

Lynx, so far as I can tell, is an attempt at implementing a text-mode HTML
browser. HTML is defined by a standard, and Lynx should therefore
implement this standard. Besides, collapsing <BR>s loses no textual
information. And seeing this behavior in a browser might incite authors to
actually read the documentation and find out what <BR> is *really* for. 

What Lynx does in cases not covered by relevant standards is often
influenced by prevailing bad HTML authoring style (e.g. the use of ../ in
URLs relative to /,) but the case of multiple <BR>s is clearly handled by
the HTML 3.2 standard.